. Author manuscript; available in PMC: 2023 Dec 23.

Published in final edited form as: J Antimicrob Chemother. 2022 Dec 23;78(1):242–251. doi: 10.1093/jac/dkac382

Table 3.

Results of the interventional ARIMA analyses

	Model 1: Hydroxychloroquine U.S.		Model 2: Hydroxychloroquine Canada		Model 3: Ivermectin U.S.		Model 4: Ivermectin Canada
ARIMA model	(2,1,0) no intercept		(4,1,1) no intercept		(2,1,0) × (0,1,0)₁₂ no intercept		(2,1,0) no intercept
Intervention:	Estimate (95% CI)	p-value	Estimate (95% CI)	p-value	Estimate (95% CI)	p-value	Estimate (95% CI)	p-value
1 (pulse)^a	117.50 (102.03, 132.97)	<0.0001	111.55 (95.60, 127.50)	<0.0001	–	–	–	–
2 (pulse)^a	–	–	–	–	0.85 (−1.17, 2.87)	0.41	0.07 (−0.03, 0.16)	0.16
3 (pulse)^a	–	–	–	–	6.16 (2.66, 9.65)	0.0006	0.44 (0.36, 0.51)	<0.0001
4 (ramp)^a	–	–	–	–	1.94 (1.11, 2.75)	<0.0001	0.02 (0.003, 0.04)	0.02

Interventions are described in Tables 1 and 2. Intervention 1 is for hydroxychloroquine, while interventions 2–4 are for ivermectin. Results that are not applicable to the model are marked with a dash (–).